C-TYPE NATRIURETIC PEPTIDE (CNP): CARDIOVASCULAR ROLES AND POTENTIAL AS A THERAPEUTIC TARGET

Abstract

Natriuretic peptides play a fundamental role in cardiovascular homeostasis by modulation of fluid and electrolyte balance and vascular tone. C-type natriuretic peptide (CNP) represents the paracrine element of the natriuretic peptide axis which complements the endocrine actions of atrial natriuretic peptide (ANP) and brain natriuretic peptide (BNP). CNP is produced by the endothelium and the heart and appears to play a prominent role in vascular and cardiac function, both physiologically and pathologically. This provides a rationale for the therapeutic potential of pharmacological interventions targeted to CNP signalling. This article provides an overview of the biology and pharmacology of CNP, with emphasis on the cardiovascular system, and discusses pathologies in which drugs designed to manipulate CNP signalling maybe of clinical benefit.

Introduction

The natriuretic peptide (NP) family comprises three principal members which are released in response to hypervolaemic and hypertensive states. Atrial natriuretic peptide (ANP), B-type natriuretic peptide (BNP), and C-type natriuretic peptide (CNP) are all involved in the maintenance of electrolyte-fluid balance and vascular tone; as their name suggests, they promote natriuresis and diuresis resulting in loss of sodium and water thereby lowering blood volume and blood pressure. CNP was the third NP to be identified, first purified in 1990¹ some 10 years after the discovery of ANP². CNP is the most widely expressed NP, with hotspots including the brain, chondrocytes and endothelial cells. CNP is thought to act locally, as a paracrine/autocrine regulator, since it is cleared rapidly from the circulation and present at very low concentrations in plasma³. It does not possess the potent diuretic actions that are observed with ANP and BNP^3-5, but it is well established that exogenous CNP is a potent arterial- and veno- dilator of isolated mammalian blood vessels^6-13 and infusion of CNP lowers blood pressure in several species, including humans^14-17. Indeed, CNP has also been characterised as an endothelium-derived hyperpolarizing factor (EDHF)⁷. CNP also exerts direct cardiac effects where it acts as an inotropic and chronotropic agent^18-20 and reduces aldosterone synthesis²¹. Chronic regulation of blood pressure by CNP also includes a central component through suppression of ACTH²¹, vasopressin²², and sympathetic outflow²³. CNP exerts a number of physiological effects outside the cardiovascular system; perhaps its primary role being regulation of bone growth^24-26, but it likely serves other functions such as neuronal development²⁷, neuroprotection²⁸, and reproduction²⁹. CNP gene (Nppc) disruption has revealed the importance of CNP in terms of bone growth (endochondral ossification). Nppc KO mice exhibit dwarfism with the length of femurs, tibiae and vertebrae being 50 – 80 % of their wild-type littermates, in addition to striking narrowing of the growth plate²⁵. Less than 50 % of CNP knockouts (KOs) are able to survive during postnatal development, although targeted expression of CNP in the growth plate chondrocytes improves their survival rate²⁴.

Synthesis, structure and regulation of CNP production

Natriuretic peptides possess a core 17aa ring structure, critical for receptor binding, which is highly conserved between members of the gene family and also amongst species. Of the natriuretic peptides, CNP is the most highly conserved^{30, 31}. The natriuretic peptide genes each encode prepropeptides and in man the CNP gene, Nppc, is located on chromosome 2 and consists of 2 exons³². Following translation, removal of the signal peptide converts the 126aa human prepro-CNP to the propeptide. In turn, pro-CNP is cleaved by furin³³, a ubiquitous proprotein convertase resident in the trans-golgi network³⁴, to yield CNP-53. This peptide possesses inherent biological activity³ but is typically processed further, by an as yet unidentified mechanism, to yield the predominant biologically active, 22aa form (CNP).

Unlike its sibling peptides, CNP is not stored and secreted from granules, but it remains unclear whether prepropeptide processing occurs ‘on demand’, particularly in terms of the endothelium-dependent vasorelaxant activity of CNP (see below)^{7, 35}. Nonetheless, there are several important stimuli that promote the secretion of CNP from endothelial cells to alter cardiovascular function including pro-inflammatory cytokines such as tumour necrosis factor and interleukin 1, bacterial lipopolysaccharide, and transforming growth factor-β^{36, 37}; CNP release from the endothelium is also triggered by shear stress akin to nitric oxide (NO)³⁸.

CNP bioactivity is tightly regulated by two principal breakdown pathways, entailing that clearance from the plasma is very rapid (t_½~3 min)³⁹. First, CNP has high affinity for the natriuretic peptide ‘clearance’ receptor (NPR-C), which promotes internalisation and lysosomal degradation of each member of the natriuretic peptide family^{40, 41}. Second, CNP is rapidly hydrolysed by neutral endopeptidase, a metalloproteinase present in the plasma and on the endothelial cell surface⁴².

Natriuretic peptide receptors underlying CNP bioactivity

CNP (and other natriuretic peptides) bring about their biological effects via activation of specific natriuretic peptide receptors (NPR), of which three have been cloned and characterised (NPR-A, NPR-B and NPR-C). The crystal structures of NPR-A and NPR-C bound to natriuretic peptide ligands have been solved, whereas similar structural information for NPR-B has yet to be published^{43, 44}. These proteins exhibit high homology as might be expected, particularly in their N-terminal extracellular ligand binding domains, and short transmembrane regions (NPRs are homodimeric)⁴⁵. NPR-A and NPR-B also comprise analogous kinase homology and dimerisation domains and, at the C-terminus, a guanylate cyclase functionality that generates the intracellular second messenger cyclic GMP for signalling purposes. In contrast, NPR-C has a short 37 amino acid intracellular tail that lacks cGMP synthesising capability but consists of overlapping G_i/o binding sequences that have been shown to trigger a number of G-protein linked processes including inhibition of adenylate cyclase activity and phospholipase-Cβ3 turnover⁴⁶. CNP has low affinity and efficacy (i.e. cGMP production) at NPR-A (in comparison to ANP and BNP) and therefore this NPR subtype is not thought to mediate physiological responses to CNP^47-49. In contrast, CNP represents the sole endogenous ligand for NPR-B and can also bind and activate NPR-C; it is therefore these NPR subtypes that mediate the bioactivity of CNP and are considered in more detail here.

NPR-B

The human NPR-B gene, Nppb, is located on chromosome 9 and consists of 22 exons producing a 1047 amino acid protein of approximately 120 kDa⁵⁰. NPR-B is expressed highly in the brain, kidney, heart and vascular tissue^{50, 51}. CNP selectively activates the receptor at physiological concentrations (pM range) and has a binding affinity some 50–500 fold higher than the other natriuretic peptides (CNP>ANP≥BNP)^{47, 49}. In NPR-B KO mice there is dramatic impairment of endochondral ossification and attenuation of longitudinal vertebra or limb-bone growth leading to reduced body size and dwarfism⁵². Using this model it has also been shown that NPR-B is important in the embryonic development of the female reproductive tract⁵². In the cardiovascular system, NPR-B is found predominantly on veins, but also in arteries. The vasorelaxant activity of CNP in larger conduit vessels is dependent on NPR-B-triggered cGMP generation, whereas in resistance arteries CNP responses are mediated predominantly via NPR-C (see below)^{11, 35}.

NPR-C

NPR-C is encoded on human chromosome 5 and is a 540 amino acid peptide⁵⁰. The mature receptor protein exists as a homodimer (each monomer having a size of approximately 60 kDa⁵³) which is stabilised by juxtamembrane intermolecular disulphide bonds⁵⁰. NPR-C is the most abundantly expressed and widely distributed NPR and is found in major endocrine glands, lungs, kidneys and the vascular wall⁵⁴ (e.g. vascular endothelial cells). All natriuretic peptides bind to NPR-C with high affinity but it has the following selectivity profile: ANP>CNP>BNP⁴⁷. The development of NPR-C KO mice has confirmed the clearance function of NPR-C⁴¹. The plasma half-life of ANP is significantly extended in these animals, although plasma levels per se are unaffected. Mice lacking NPR-C have increased basal bone turnover and skeletal abnormalities⁴¹, which is thought to arise from reduced clearance of CNP that augments bone growth indirectly via NPR-B.

CNP as an EDHF

In addition to the clearance function of NPR-C, increasing evidence intimates its importance in regulating blood vessel tone in the resistance vasculature. Endothelium-derived hyperpolarising factor (EDHF) is a term used to describe the mediator(s) responsible for endothelium-dependent relaxation that is non-NO and non-prostanoid in nature, and which brings about hyperpolarisation of vascular smooth muscle via opening of inwardly-rectifying potassium channels (K_ir) and/or activation of a Na⁺/K⁺-ATPase⁵⁵. To date, there is no consensus on the identify of EDHF, although several candidates have been proposed including electrical coupling via myoendothelial gap junctions, hydrogen peroxide and cytochrome P450 metabolites⁵⁵. Evidence from our lab has also added CNP to this list of putative EDHFs^{7, 35, 56}. Furthermore, the EDHF-like actions of CNP have been linked to activation of NPR-C since the selective antagonist for this receptor, M372049, attenuates EDHF responses in rat mesenteric arteries³⁵ and the selective NPR-C agonist cANF^4-23 mimics the EDHF activity of CNP. As a result, it has been proposed that NPR-C- and G_i-dependent activation of an inwardly rectifying potassium channel (so called G-protein coupled K_ir channels or GIRKs) underlies this activity. Human studies also support an EDHF-like function of CNP; in forearm resistance vessels, CNP-induced vasodilatation is dependent on hyperpolarisation and accentuated by the neutral endopeptidase inhibitor thiorphan¹⁶. Moreover, in human penile resistance arteries CNP has been shown to display EDHF-like activity via an NPR-C-dependent mechanism⁵⁷. Importantly however, the vascular functions of endothelium-derived CNP do not seem to end with regulation of vascular tone. CNP has also been shown to govern platelet and leukocyte reactivity⁵⁸, endothelial and vascular smooth muscle cell growth^{59, 60}, and cardiac hypertrophy and ischaemia-reperfusion injury^{56, 61}; many of these functions appear to be regulated by NPR-C. These additional biological effects are important when considering the potential for therapeutic modulation of CNP/NPR-C signalling, since this multi-faceted profile of activity brings about a multi-pronged cytoprotective effect in the vasculature (Figure 1); this is discussed further below.

Figure 1.

Schematic depicting the multi-faceted, anti-atherogenic profile of endothelium-derived CNP

Therapeutic potential of interventions targeted to CNP signalling

Hypertension

Hypertension is a significant public health issue in the developed world and its prevalence continues to increase; nearly 100 million adults in the United States are thought to suffer from the disease^62-64. Moreover, as many as 30% of hypertensive patients are known to be resistant to medication regardless of compliance, thus highlighting a need for the development of novel therapies⁶⁵. Hypertension predisposes individuals to myocardial infarction, stroke, renal failure and other cardiovascular diseases⁶⁶ that account for almost one third of total global deaths (WHO statistics).

Hypertension is associated with aberrant endothelial function, including loss of NO bioactivity, and increased shear stress resulting in vascular remodelling and dysfunction⁶⁷. Considering that CNP secretion is known to be augmented in response to shear stress^{38, 68} and that it exerts anti-remodelling effects on vascular smooth muscle and endothelial cells^69-72, it is plausible that endothelial CNP is protective against vascular insults associated with hypertension and pharmacological manipulation of the peptide may be of therapeutic benefit. Unfortunately, rigorous functional pharmacological studies investigating the contribution of CNP-NPR-B/NPR-C in regulating blood pressure have been hampered by the fact that genetic deletion of CNP, NPR-B or NPR-C (in mice) results in gross bone deformation and premature death^{24, 41, 52}. Nonetheless, genetic data do highlight a link between CNP (and/or NPR-C) signalling and hypertension in humans. For instance, a polymorphism in the 3′-UTR region of the Nppc gene is associated with hypertension in a Japanese population⁷³. Additionally, a significant increase in circulating CNP correlates positively with hypertension in patients requiring catheterisation for coronary angiography⁷⁴. Furthermore, allelic variants in the 3′-UTR of NPR-A and NPR-C have been linked to familial hypertension and higher systolic blood pressure⁷⁵ and the number of tandem repeats in the 5′-flanking region of NPR-C correlates with increased mean arterial blood pressure⁷⁶. Pharmacological evidence also predicts that targeting (i.e. augmenting) CNP signalling may be clinically useful. Administration of exogenous CNP causes an acute drop in blood pressure in primates¹⁴, dogs¹⁵ and humans^{15, 17}, and spontaneously hypertensive rats⁷⁷. CNP also exerts chronic regulation of blood pressure through suppression of ACTH²¹, vasopressin²², aldosterone synthesis²¹ and sympathetic outflow²³. CNP may also play a protective role in hypertension via NPR-B activation; CNP and NPR-B mRNA levels are increased in the aorta of obese hypertensive rats when compared to lean rats and the NPR-A/B antagonist, HS-142-1, increases systolic blood pressure in obese rats that is accompanied by a decrease in cGMP production⁷⁸.

Myocardial infarction and cardiac remodelling

Restoration of blood flow, or reperfusion, of ischemic tissues is essential for limiting and salvaging organ function in many cardiovascular disorders, including myocardial infarction (MI). However, reperfusion per se exerts detrimental effects by promoting local inflammation, triggering cell death and facilitating microvascular dysfunction; this phenomenon is known as ischemia/reperfusion (I/R) injury⁷⁹. The inflammation and vascular dysfunction associated with I/R injury are underpinned by loss of endothelium-derived mediators including nitric oxide (NO); this results in vasoconstriction, decreased perfusion, leukocyte recruitment and platelet activation, which combine to exacerbate damage.

Endogenous and exogenous CNP may offer protection from I/R injury. Ventricular hypertrophy, necrosis, inflammation and functional impairment of cardiac tissue induced by coronary artery ligation is reduced in mice over expressing CNP in cardiomyocytes⁶¹. In accord, chronic CNP infusion administered post MI attenuates cardiac fibrosis and left ventricular hypertrophy, and increases survival⁸⁰. The beneficial effects of CNP (particularly anti-hypertrophic activity) in this scenario may be due to NPR-B activation since transgenic rats over expressing a dominant negative NPR-B exhibit decreased cGMP production and cardiac hypertrophy⁸¹. Such a thesis is supported by the observation that CNP down-regulates the expression of many pro-hypertrophic genes/transcription factors, including myocyte enhancer factor 2 (MEF2) and GATA4, and opposes the hypertrophic effects of endothelin-1 and angiotensin II^82-84. CNP signalling (perhaps via NPR-C) is also believed to regulate the pacemaker current in the heart⁸⁵. Furthermore, CNP or drugs mimicking its bioactivity may be beneficial in the treatment of acute MI. Our work has shown that CNP and cANF^4-23 reduced infarct size following I/R injury in the rat isolated heart⁵⁶. Here, NPR-C activation seems to underlie the salutary effects of CNP on the myocardium and in regulating coronary blood flow. Additionally, the activity of CNP is enhanced in the absence of endogenous NO production indicating that CNP may play a compensatory role in protecting the heart and vasculature when NO signalling is impaired.

Angiogenesis & revascularisation

One process thought to play a key role in protecting against I/R injury is re-vascularisation, the growth and development of new blood vessels to re-establish blood flow to the ischaemic tissue. A key facet of this process is angiogenesis, an endothelium-centred response to hypoxia that promotes the migration and proliferation of differentiated endothelial cells to sprout new capillaries from existing blood vessels. This is initiated predominantly by hypoxia inducible factor (HIF)1α that in turn transcriptionally-activates genes including vascular endothelial growth factor (VEGF), VEGF receptors (i.e. Flt-1, Flk-1/KDR & neuropilin-1), angiopoietins, fibroblast growth factor (FGF), platelet-derived growth factor (PDGF) and cytokines such as IL-6^86-88.

It is now clear that the endothelium plays a pivotal role in integrating re-vascularisation pathways to expedite tissue repair and minimise functional deficit following ischaemia. Of the endothelium-borne signalling molecules that govern re-vascularisation, it is arguably NO that has been best-characterised^{87, 89}. Despite the elucidation of several NO-dependent mechanisms that lead to re-vascularisation, little is known about additional endothelium-derived mediators that may contribute to this process.

Interestingly, several reports in the literature suggest that endothelium-derived CNP can mobilise endothelial cells and promote re-endothelialisation of damaged blood vessels, consistent with a pro-angiogenic activity. For example, adenoviral delivery of CNP slows neointimal hyperplasia and promotes re-endothelialisation in vein grafts⁶⁹ and following balloon angioplasty^{60, 70-72, 90}, and maintains blood flow and capillary density after hind-limb ischaemia⁹¹. This is fitting since many of the primary stimuli for CNP release from endothelial cells, including shear stress^{38, 68}, transforming growth factor (TGF)β⁹², VEGF and monocyte chemoattractant protein (MCP)-1^{93, 94}, are key to the re-vascularisation process and should enlist CNP bioactivity. Further, systemic infusion of CNP enhances endothelial growth and prevents development of intimal thickening in damaged carotid arteries⁵⁹, and CNP expression is highly up-regulated in human coronary atherosclerotic lesions⁹⁵ and in the neointima following percutaneous coronary intervention (PCI)⁹⁶. These observations all intimate that CNP is produced endogenously in response to vessel injury and may represent a further, key endothelium-derived mediator involved in re-vascularisation and a promising therapeutic target.

Heart Failure

Heart failure is defined as an impaired ability of the ventricle to fill with or eject blood, and is characterized by symptoms including dyspnoea and fatigue. It is a serious and growing health concern, with mortality increasing despite advances in treatment. In the USA nearly 5 million people suffer from heart failure, and there are more than 50,000 deaths per annum⁹⁷.

The development of heart failure involves abnormal signalling in a number of pathways, but particularly those stimulating cGMP production; this encompasses aberrant natriuretic peptide (particulate guanylate cyclase) and NO-mediated (soluble guanylate cyclase) signalling^{98, 99}. NO-donor drugs restore NO levels in heart failure and have been used to treat the disease for over a century, but their long-term use is limited by the development of tolerance and cGMP-independent side effects¹⁰⁰. Thus, activation of particulate guanylate cyclases offers a promising alternative. Recently, chimeric natriuretic peptides have been developed that appear to harness many of the desirable biological activities of each natriuretic peptide in heart failure^{101, 102}. This approach is exemplified by the development and clinical evaluation of CD-NP, a chimera of the C-terminus of Dendroaspis natriuretic peptide (DNP) fused to CNP, which reproduces the beneficial renal effects of BNP and vasodilatory actions of CNP without the excessive hypotensive effects that are often observed with treatment of heart failure patients with Carperitide and Nesiritide (synthetic ANP and BNP, respectively). Heart failure is one of the few conditions that results in an increase in plasma CNP levels, so it may be that CNP acts in both a paracrine (endothelial dysfunction is characteristic of heart failure) and endocrine fashion (cardiac function) in this setting. Moreover, cardiac production of CNP and expression of NPR-B (more so than NPR-A) are increased in heart failure, suggesting it is released as a cytoprotective mechanism^103-106.

Atherosclerosis

Atherosclerosis is a multi-factorial pathological and inflammatory process of the blood vessel wall leading to distal ischaemic events such as MI and stroke. Development of atherosclerosis begins early in life¹⁰⁷ and is a major risk factor for ischaemic cardiovascular diseases which are the primary causes of death in the western world^{108, 109}. Initiation of atherosclerosis is influenced by a number of ‘risk factors’ including smoking, hypertension, hyperlipidaemia, and diabetes mellitus, all of which cause endothelial dysfunction¹¹⁰. Atherosclerotic plaques form in arterial sites where there is decreased shear stress^{111, 112}; in a healthy vascular system shear stress results in a gene expression profile which is crucial for normal vascular function and is atheroprotective by inhibiting proliferation, thrombosis and inflammation of the vessel wall^{113, 114}. Normal laminar flow enhances the expression and phosphorylation (i.e. activation) of eNOS and the production of NO, which promotes vasorelaxation, inhibits platelet activation, and monocyte adhesion^115-118. At curvatures, branch ostia and bifurcations, there is disruption to the blood flow and hence reduced shear stress, which is associated with reduced eNOS production, an increase in reactive oxygen species (ROS), leukocyte adhesion, smooth muscle cell proliferation and collagen deposition^{113, 114}. Reduction in shear stress causes up-regulation of endothelial and leukocyte adhesion molecules including selectins, integrins, intercellular adhesion molecule-1, and vascular-cell adhesion molecule-1¹¹¹. Expression of these molecules on the endothelial surface causes recruitment of monocytes and T-cells and the migration of leukocytes into the artery wall. It is the macrophages in the artery wall that take up oxidised-LDL to become foam cells. The presence of foam cells and T lymphocytes marks the beginning of the fatty streak formation¹¹¹. The immune cells release a milieu of inflammatory mediators including growth factors and cytokines, which cause migration and proliferation of the underlying smooth muscle cells, which can then release their own mediators including IL-1β, TNF-α and β, IL-6, macrophage colony-stimulating factor, MCP-1, IL-18 and CD-40L resulting in a self-perpetuating inflammatory process within the plaque. As the lesion grows it forms a fibrous cap (often associated with calcification) that isolates the lesion from the lumen and this can become progressively weaker resulting in plaque rupture, thrombosis and distal occlusion.

CNP has been shown to possess anti-atherogenic properties. Immunological studies on human coronary atherosclerotic lesions have revealed that CNP is present in different cell types according to progression of atherosclerosis. Under normal physiological conditions, arterial segments show CNP-positive endothelial cells⁹⁵ and monocytes/macrophages¹¹⁹. As the development of the atherosclerotic plaques progresses, advanced lesions have CNP-positive macrophages but the majority of the smooth muscle cells within the fibrous cap and the surface endothelial cells are CNP-negative⁹⁵. Further still, CNP mRNA levels are decreased in human stenosed aortic valves¹²⁰. In sum, such observations intimate that the severity of atherosclerotic lesions inversely correlates with CNP expression¹²¹. It is also interesting to note that oxidized LDL suppresses the secretion of CNP from endothelial cells¹²². CNP has also been implicated in the regulation of leukocyte adhesion, platelet activation and P-selectin expression⁵⁸. CNP is able to suppress basal and IL-1β-induced leukocyte recruitment and this is mimicked by the NPR-C selective agonist cANF^4-23, implying the involvement of NPR-C. CNP also prevents thrombin-induced platelet aggregation. It is also worth noting that relaxation to CNP in atherosclerotic aortae from cholesterol rabbits is impaired compared to aortae from normal dietary fed rabbits; however, incubation of the atherosclerotic aortae from cholesterol rabbits with a neutral endopeptidase inhibitor attenuates the impairment of the vasorelaxation response to CNP¹²³.

Taken together, the above evidence suggests an anti-atherogenic role for CNP by regulating smooth muscle migration, cellular adhesion and P-selectin expression. Further evidence supporting a cytoprotective role for CNP is provided by studies with vascular endothelial cells which show that the switch from a reductive to an oxidative environment in atherosclerosis (and the production of reactive oxygen species such as superoxide and hydrogen peroxide), and key mediators in inflammation and vascular remodelling augment endothelial secretion of CNP^{36, 37, 124}. Moreover, diabetes associated atherosclerosis might be in part due to impairment of CNP since insulin suppresses CNP, but not endothelin-1, production by cultured endothelial cells¹²⁵. Interestingly, CNP attenuates the inhibitory effect of Angiotensin-II on Gax (Growth arrest-specific homeobox) mRNA expression in rat aortic vascular smooth muscle cells¹²⁶. Since Gax over-expression has been reported to reduce neointimal thickening post balloon injury in the rat carotid artery¹²⁷, Gax may be a pivotal transcription factor involved in the anti-proliferative and anti-remodelling activity of CNP in vascular smooth muscle cells.

Restenosis & vein graft disease

Restenosis describes the recurrence of intimal hyperplasia that is often associated with percutaneous coronary interventions (PCI; balloon angioplasty) and stent insertions in patients with diseased and occluded coronary arteries. It is a significant clinical problem limiting the long term therapeutic success of these procedures^{128, 129}. Drug eluting stents (DES) release agents with anti-proliferative and immunosuppressive actions (e.g. sirolimus/rapamycin) which have been successful in reducing the rate of restenosis to <10%¹³⁰. However, the complication of stent thrombosis remains¹²⁹. Stent thrombosis post PCI is categorised as early (0-30 days post PCI), late (>30 days to 1 year post PCI) or very late (>1 year post PCI)¹³¹. Stent thrombosis occurs in patients treated with bare metal stents and DES but the rate of late stent thrombosis is almost doubled in patients treated with DES¹³². Agents released by DES inhibit vascular smooth muscle proliferation and migration, a beneficial action in reducing the incidence of restenosis; however they also inhibit endothelial cell proliferation¹³³, an undesirable effect. PCI and stent deployment cause endothelial denudation, exposing thrombogenic material to the blood, which can lead to thrombosis^{134, 135}. Therefore, an agent that could concomitantly prevent vascular smooth muscle hyperplasia but promote endothelial proliferation should offer a considerable therapeutic advance.

CNP appears an ideal candidate to be released from DES as it promotes endothelial cell proliferation, inhibits VSMC proliferation, and has anti-inflammatory and anti-thrombotic properties. In a porcine model of femoral artery restenosis post balloon angioplasty, liposome-mediated CNP gene and peptide administration reduces neointima formation without compromising re-endothelialisation⁷¹. In balloon injured rabbit carotid arteries infection with an adenovirus expressing human CNP leads to increased NO production and suppression of neointimal thickening, ICAM-1 and VCAM-1 expression, and macrophage infiltration¹³⁶. This profile of activity is also observed following angioplasty in a number of further vessels and species, including rabbit carotid artery, rabbit femoral artery and porcine femoral artery^{90, 136-138}.

A similar beneficial effect of CNP is reported in vein grafts, often used for auto transplantation in patients requiring coronary artery bypass surgery. Akin to (coronary) artery stenosis, following a vein graft procedure the transplanted vessel is affected by thrombosis, intimal hyperplasia and atherosclerosis¹³⁹. Within the first month, approximately 13% of saphenous vein grafts occlude, principally due to thrombosis^{140, 141}, and the 10 year vein graft patency is ~ 50%^{141, 142}. CNP has been shown to reduce intimal hyperplasia and CD-8 positive cells in a mouse model of vein graft disease, in which the inferior vena cava is interposed into the common carotid artery¹⁴³. Furthermore, CNP promotes re-endothelialisation and suppresses neointimal hyperplasia and thrombosis in rabbit jugular vein grafts interposed into the carotid artery⁶⁹. Such studies clearly demonstrate CNP to be effective in suppressing thrombosis and neointimal hyperplasia, two of the main processes involved in vein graft disease.

Pulmonary hypertension

Pulmonary hypertension (PH) is a debilitating disease with a poor prognosis. It is characterised by increased pulmonary artery pressure, re-modelling of the pulmonary resistance vasculature right heart hypertrophy, eventually leading to right heart failure and death^144-146. Current therapeutic options include prostacyclin analogues^{147, 148} endothelin receptor antagonists^{149, 150}, and PDE5 inhibitors¹⁵¹, but despite these advances 2-year mortality remains unacceptably high (~15%)¹⁵². Thus, PH represents a clear unmet medical need.

In PH, the bioactivity of NO and other endothelium-dependent vasodilators such as prostacyclin is reduced^{145, 153, 154}, whereas production of endogenous vasoconstrictors such as ET-1 is increased¹⁵⁵. This pulmonary endothelial dysfunction precipitates the excessive pulmonary vasoconstriction and remodelling that characterise the disease. The strategy of elevating cGMP by augmenting NO-dependent signalling (e.g. NO inhalation, direct sGC stimulation, PDE5 inhibitors), is clinically effective in PH¹⁵⁶. We and others have recently reported that pharmacological interventions targeted to the natriuretic peptide pathways (which also results in augmented cGMP production) may be an alternative and more efficacious approach to the treatment of PH^{157, 158}. Inhibitors of neutral endopeptidase augment natriuretic peptide bioactivity and produce a selective pulmonary dilatation; this mechanism also leads to a reduction in pulmonary vascular remodelling and right ventricular hypertrophy¹⁵⁷. Thus, a precedent for a natriuretic peptide-centric approach to PH has been established. As such, altering CNP signalling may prove therapeutically effective. This idea is corroborated by the report that CNP ameliorates the development of PH in an animal model¹⁵⁹ and that circulating levels of CNP are increased by hypoxia¹⁶⁰. In addition, the anti-remodelling (vascular & cardiac) and anti-inflammatory actions of CNP, coupled with its ability to promote re-endothelialisation (above), intimate it should be protective in PH.

Conclusions

A key role for CNP in regulating cardiac and vascular function has slowly emerged since its identification 20 years ago^{161, 162}. This cardiovascular homeostatic role is perhaps best exemplified by the ability of CNP to control blood flow in the resistance vasculature⁷, to reduce the reactivity of leukocytes and platelets⁵⁸, to govern the pacemaker current in the heart⁸⁵, to protect against myocardial I/R injury^{56, 80} and cardiac hypertrophy^{61, 82}. Moreover, elevated plasma CNP levels are found in patients with heart failure^{106, 163}, septic shock¹⁶⁴ and renal dysfunction¹⁶⁵, thereby associating this endothelium-derived peptide with human cardiovascular pathology. The multi-faceted cytoprotective functions assigned to CNP described herein suggest that pharmacological manipulation of this natriuretic peptide, and its receptors (both NPR-B and NPR-C), hold great promise in advancing the treatment of cardiovascular disease. This has already begun in many respects with the clinical evaluation and use of neutral endopeptidase inhibitors¹⁶⁶ and CD-NP¹⁰², and as our understanding of the cardioprotective roles of CNP expands, so this interest should grow.